Huaisong Wu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Huaisong Wu has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Bin Luo
College of Mechanical and Electrical Engineering, Beijing University of Chemical Technology, Beijing 100029, China
1 shared publication - Bingyang Li
China Academy of Aerospace Science and Innovation, Beijing 100871, China
1 shared publication - Futao Wang
Qian Xuesen Laboratory of Space Technology, China Academy of Space Technology, Beijing 100094, China
1 shared publication - Pengfei Wang
China Academy of Aerospace Science and Technology Innovation, Beijing 100176, China
1 shared publication - Qi Zhang
Beijing University of Chemical Technology
1 shared publication - Yu Yuan
Control Theory and Engineering, School of Astronautics, Harbin Institute of Technology, Harbin, China
1 shared publication
Latest Publications
- Open Access
- 12 November 2021
- Published by MDPI AG in Applied Sciences
- Vol. 11 (22) , 10667
Design and Analysis of the Rolling and Jumping Compound Motion Robot
Complex and unknown areas in deep space exploration present major challenges to the motion ability of current space robots. Different from the traditional single-mode motion space robot, a compound motion robot with flexible movement and strong obstacle surmounting ability is proposed. Through the highly integrated structure design, the lightweight robot has the ability of rolling and jumping, and the kinematic characteristics of the robot under two motion modes are analyzed. This work provides a reference for the design of deep space exploration equipment with high motion capability in the future.
- 12 November 2021
- Published by MDPI AG in Applied Sciences
- Vol. 11 (22) , 10667
Open AccessDesign and Analysis of the Rolling and Jumping Compound Motion Robot
Complex and unknown areas in deep space exploration present major challenges to the motion ability of current space robots. Different from the traditional single-mode motion space robot, a compound motion robot with flexible movement and strong obstacle surmounting ability is proposed. Through the highly integrated structure design, the lightweight robot has the ability of rolling and jumping, and the kinematic characteristics of the robot under two motion modes are analyzed. This work provides a reference for the design of deep space exploration equipment with high motion capability in the future.
- 1 August 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 684-687
License Plate Location Based on Combination of Deep Learning and Feature Fusion
License plate location is the key step of license plate recognition. In order to improve the precision of license plate location, a license plate location method based on deep learning and feature fusion is proposed in this paper. Firstly, the color feature of license plate is extracted by histogram backprojection method, and the texture feature of license plate is extracted by Sobel Operator. Combining these two kinds of features, the obtained connectivity regions are used as the candidate regions of license plate. And then ResNet20 convolutional neural network is used to identify the candidate regions and locate the license plate. The test results show that the method can satisfy the precise license plate location under the different shooting angles and illuminations, and lay a foundation for improving the accuracy and robustness of license plate recognition in the future.
- 1 August 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 684-687
License Plate Location Based on Combination of Deep Learning and Feature Fusion
License plate location is the key step of license plate recognition. In order to improve the precision of license plate location, a license plate location method based on deep learning and feature fusion is proposed in this paper. Firstly, the color feature of license plate is extracted by histogram backprojection method, and the texture feature of license plate is extracted by Sobel Operator. Combining these two kinds of features, the obtained connectivity regions are used as the candidate regions of license plate. And then ResNet20 convolutional neural network is used to identify the candidate regions and locate the license plate. The test results show that the method can satisfy the precise license plate location under the different shooting angles and illuminations, and lay a foundation for improving the accuracy and robustness of license plate recognition in the future.